2. Consider the data in Exercise 8 in Section 10.2 on fixed-platen compression strengths of three types of corrugated fiberboard containers.

(a) One of the assumptions needed for the validity of the ANOVA F test is homoscedasticity, or equal variances.

Since only summary statistics are given, the homoscedasticity assumption can be tested by performing all pairwise tests for the three population variances and use the Bonferroni multiple comparisons procedure. In particular, apply the F test for the equality of two variances, given in Section 9.4, to test H10 : σ2 1 = σ2 2 , H20 : σ2 1 = σ2 3, and H30 :

σ2 2 = σ2 3 versus the corresponding two-sided alternative hypotheses, and apply the Bonferroni method at experiment-wise level of significance 0.1. Does the homoscedasticity assumption appear to be violated?

(b) Use the Bonferroni multiple comparisons method, with pairwise two-sample T tests without the equalvariance assumption, to identify the types of corrugated containers whose population means differ significantly at experiment-wise level of significance

α = 0.05. Summarize your results by arranging the means in increasing order and underlining the ones that do not differ significantly.